ON LANGUAGE NAMES /// RE: Results of Duplicate Busters Survey #2

Lang Gérard gerard.lang at insee.fr
Mon Sep 8 12:22:31 CEST 2008

Dear Peter Constable,

As a professional statistician, as well as a specialist of information systems, I consider that your reaction regarding distinction between the concept of language name and the concept of language as merely a philosophical question having nothing to do with language tags and  is completely false.
Language names code elements and languages tags are second order concepts, that are "concrete" realisations of the higher order concepts of language name and language.
Any misconception about the higher order concepts is practically certain to call misapplications, when coming to language names coding elements and language tagging. 
The two essential proprieties that are called in this question are:
1-Not to give the same language tag to two distinct language, to insure that IETF language tags are correctly managed;
2-So, not to give the same language name code element to two distincts language names is a precondition to insure that correct management, so that the two underlying languages are distincly named;
3-It would also be much useful that two distinct language names, having possibly the same underlying language, should not be recognized as two distinct entries inside ISO 639, so as not to have competition between two code elements for the choice of a given language tag. But this result seems much more delicate to reach , and this explains why I am very much advocating that autonyms should be primarily retained as language names, not french or english language names, because they much more separating and identyfying power and unicity.

Gérard LANG

-----Message d'origine-----
De : Peter Constable [mailto:petercon at microsoft.com] 
Envoyé : vendredi 5 septembre 2008 18:29
À : Lang Gérard; Michael Everson; ietflang IETF Languages Discussion
Objet : RE: ON LANGUAGE NAMES /// RE: Results of Duplicate Busters Survey #2

From: ietf-languages-bounces at alvestrand.no [mailto:ietf-languages-bounces at alvestrand.no] On Behalf Of Lang Gérard

> 1-It is not true that no one else is caring much on these questions, 
> because I am very much caring about the question of language names.
> 2-Let me remind you that ISO 639 generic title is "Codes for the 
> representation of language names"...

This basically amounts to a philosophic discussion as to what the ISO 639 standards are or should be all about and how they should work. An interesting discussion, but I think not one that's on topic for this list.

Here, we are presented with a family of standards that, factually, publish certain names, and we need to determine how best to use that input in selecting names that we record for ISO 639 language IDs in the Language Subtag Registry.

While this philosophic discussion has not taken place either in this list or in the LTRU working group, I would venture to guess that there is a general consensus here that IETF language tags are considered to be symbolic, coded elements that denote concepts, those concepts being one of the following:

- the concept of a language -- i.e., some range of language varieties conventionally deemed to be a distinct language identity;

- the concept of a language variant -- i.e., one particular variety, or a sub-range of language varieties conventionally recognized as a distinct linguistic identity;

- the concept of a writing system -- i.e., a particular language or language variant in written form using some particular writing conventions involving one or more particular scripts;

- the concept of an orthography (a writing system constrained by particular spelling conventions).

Likewise, I'd guess the general understanding of language subtags derived from ISO 639 is that they are symbolic, coded elements that denote the concepts of particular individual languages. (Note I'm talking here about LSTR language subtags, not elements defined in ISO 639.) Accordingly, the purpose of names recorded in the LSTR is to document the concept denoted by a given subtag. We are tacitly assuming that the various parts of ISO 639 do provide a representation of concepts, whether that be directly or indirectly -- that is, if (say) formally ISO 639-1 is coding language names, we assume here that it is, by doing so, indirectly defining symbols that denote the concepts of particular languages.

At least, that's basically how I understand what the IETF LSTR is all about.


More information about the Ietf-languages mailing list